Authoritative Practices and Collective Validation: Wikidata within the Collaborative Digital Edition of the Greek Anthology

Maxime Guénette

Université de Montréal

Mathilde Verstraete

Université de Montréal

May 6, 2025

Introduction @àdécider

Hello, we are happy to be here
Not a surprise, Wikidata is now a commonly used tool in the DH community
(Parallèles) As DH, Wikidata functions as a knowledge base, enabling a sort of collective verification and semantic linking of information.

But somehow, that creates a conflict: Unlike traditional academic publishing, where authority is centralised and often restricted to established institutions or recognised experts, Wikidata operates through a model of continuous, multilingual, and community-based editing that promotes the dissemination of free and accessible knowledge globally. This paradigmatic shift invites a fundamental rethinking of authority, editorial responsibility, and the epistemological foundations of data.

That brings us to the question with which we will work with today:

How, then, can expert-led projects — whether developed by academics, government agencies or GLAM institutions — work with a generalist platform such as Wikidata to generate new forms of knowledge? How do these hybrid models, which combine scholarly expertise with public participation, challenge traditional boundaries between academic and amateur contributors, and between knowledge production and validation?

In the next 20 minutes, we will :
- start by exploring how the infrastructure and logic of Wikidata can be integrated into DH projects
- talk a bit about the Anthologia Graeca project, a collaborative digital edition of the Greek Anthology
- We will analyse how Wikidata is embedded in the project’s data model — particularly through our treatment of the keyword “authors” — and how this integration opens new spaces for knowledge production.
- Finally, drawing on those concrete examples, we will reflect on the tensions between authority and collective intelligence that shape such initiatives, and the ways in which digital infrastructures can both challenge and extend traditional scholarly practices.

Wikidata and Digital Humanities @maxime

Wikidata is a major knowledge graph for structuring and sharing data
DH projects use it to publish Linked Open Data without technical barriers
GLAM institutions rely on it for metadata curation and interoperability

Wikidata as A Linking Hub @maxime

One of Wikidata’s major strengths, widely recognized by Digital Humanities projects and GLAM institutions, lies in its role as a central hub for linking heterogeneous datasets. This is made possible through external identifier properties, which connect Wikidata items to corresponding entities across databases.

Take Megara, the ancient Greek city, as an example. Its Wikidata item links to a variety of sources — from library catalogues to specialized databases like Pleiades, ToposText, and MANTO. These links enable cross-referencing, facilitate authority control, and support the enrichment of data through Linked Open Data (LOD). By relying on Wikidata’s Q-items, researchers can efficiently integrate and query related resources across platforms.

Some scholars argue that Wikidata should go further and become the reference identifier itself. Multiple identifiers for the same entity — such as Megara — risk fragmenting data and complicating reconciliation. Using Wikidata as a standard could offer a unified data model, a single SPARQL endpoint, and long-term infrastructure for querying and storage.

Though this position is more ambitious, it underlines an essential aspect of Wikidata’s ecosystem: reciprocal contribution. DH projects and GLAM institutions can both feed into and benefit from Wikidata, ultimately improving the quality and reach of their data.

Authority on Digital Platforms @maxime

This brings us to a central question: how can academic projects and Wikidata mutually enrich one another, given their different systems of validation and authority? While many Wikidata editors have strong expertise, the platform is built on openness, collaborative editing, and peer consensus — rather than formal academic credentials. By contrast, academia tends to associate authority with institutional affiliation, scholarly output, and disciplinary recognition. Similarly, GLAM institutions or government agencies are seen as authoritative due to their structured oversight and perceived reliability.

These differing frameworks create asymmetries in how contributions are trusted. Academia often relies on top-down models of validation, while Wikidata operates through distributed consensus. Understanding these dynamics is essential for building stronger collaboration between both ecosystems.

We argue that traditional hierarchies are not always the most effective model for digital knowledge production. Collaborative platforms like Wikidata challenge long-standing assumptions about authority by enabling anyone to contribute. This openness raises concerns about data quality, and Wikidata editors are often viewed as amateurs or “citizen scientists.” Their work is frequently only legitimised when endorsed by academic or institutional figures.

Despite these tensions, Wikidata is increasingly recognised as a high-quality knowledge graph. Its reliability is context-dependent and must be assessed case by case. Its community plays a key role in maintaining quality, supported by tools like Shape Expressions (ShEx) that enforce data model consistency. These community-driven mechanisms represent a decentralised but structured approach to data governance.

Crucially, improving Wikidata is not just a matter of external assessment but of active engagement. Researchers, GLAM professionals, and contributors from all backgrounds can help shape the platform by adding statements, correcting errors, refining data models, and discussing ontologies. Wikidata is not a finished product but a dynamic infrastructure where quality emerges through interaction.

To rethink authority in Digital Humanities, we suggest using Wikidata not just as a reference source but as a foundational layer for data modelling, curation, and publication. In this model, epistemic authority becomes a shared process — emerging from collaboration rather than imposed from above. Academic researchers contribute alongside volunteers, curators, and technologists in shaping meaning, structure, and trust within the data.

To explore this further, we now turn to our case study: the AG project. It demonstrates how Wikidata can support participatory infrastructure and collective validation, offering a model for reshaping authority and collaboration within academic knowledge production.

A collaborative and digital edition of the Greek Anthology @mathilde

Présenter le projet
Présenter l’historique des plateformes
Parler rapidement des partenaires éventuellement (ou pas)
Keyworkds dans l’AG

The Authors of the Greek Anthology @mathilde

On va revenir sur la réalisation d’un chantier qui a été particulièrement long: celui du “nettoyage” des auteurs présents jusqu’alors sur la plateforme.
Pour la plupart, assez facile, pour d’autres, beaucoup moins :

Phanias @mathilde

Diodoros @mathilde

Dionysios @mathilde

Archias @mathilde

Marcello voulait aussi aller plus loin en ajoutant les floruit des auteurs etc, ce que j’ai refusé (manque de sources) – Maxime a dit qu’on pouvait utiliser les références (dont j’admets ne pas avoir connaissance à l’époque, oups) -> à voir, un jour peut-être
Wikidata utilisé dès lors pour des connaissances qui sont relativement fragmentaires à l’image de notre objet d’étude

Some comments on the outcome and future work @mathilde

Faire la même chose pour les autres types de mots-clés (villes – presque fini) et les personnages, les thèmes, etc. ce qui fait émerger de nouvelles questions méthodologique etc (donner les exemples)
travail en cours, on est heureux d’avoir vos retours dessus et on veillera à les intégrer à notre pratique

Conclusion @maxime

In conclusion, Wikidata is not just a technical tool — it is a collaborative epistemic space where academic knowledge is shaped, revised, and legitimised. In the AG project, we delegate part of our curatorial authority to Wikidata, embracing a shift toward distributed models of knowledge production. This does not weaken academic rigour but redefines expertise through collective intelligence and the potential of community-driven platforms to support high-quality, multilingual, interoperable data.

This shift challenges researchers to act not only as content producers but also as facilitators of open knowledge systems. It raises key questions: Can academic standards align with Wikidata’s norms? How do we balance openness with accuracy?

Rather than resolving these tensions, integrating Wikidata foregrounds them — inviting reflection on validation, authorship, and authority. Embracing such platforms means acknowledging that scholarship is always provisional, situated, and co-produced. Thank you!